We’re looking for a Senior ML Engineer (Computer Vision & Perception) to build the in-cabin Occupant Monitoring System (OMS) for our robotaxis from the ground up. As our first hire in this area, you will shape how our vehicles understand what happens inside the cabin: who is on board, whether seatbelts are fastened, whether anything was left behind, and whether passengers behave safely. You will define the approach and take the system from first prototype to real-time inference on in-vehicle hardware, working alongside experienced engineers across the company.

You will be based on site, with regular hands-on work in our vehicles, collaborating closely with our engineering teams in an agile setup. As this is a new area for us, you will help select the frameworks, tooling, and in-vehicle compute for our perception stack, with a focus on real-time performance and reliability for safety-critical functions. Day to day, we use Azure, Jira, and Microsoft Teams, and you’re free to choose your own development tools.

Who are we?

At Verne, we're building an ecosystem to progress our society through a journey of curiosity and innovation. Named in honor of Jules Verne "the man who invented the future,". We are the first European solution for autonomous journeys. We adopt an integrated approach, including vehicle design, infrastructure development, and the full digital experience.

As a growing startup, we thrive on curiosity, innovation, and bold ideas. If you’re ready to make your mark in a fast-paced, dynamic environment, this could be the role for you.

What will you do?

As an Senior ML Engineer, your responsibilities will include:

• Own the OMS perception system end to end and define its technical approach, tooling, and roadmap, from research and prototyping through to in-vehicle deployment

• Design, train, and fine-tune models for in-cabin object detection, pose estimation, and action recognition (e.g. occupancy, child presence, seatbelt use, left-behind objects, smoking, vandalism)

• Set up in-cabin data collection with the relevant teams, including camera selection and placement, and build pipelines for collecting, labelling, and curating RGB and infrared data, including staged edge cases

• Define evaluation metrics and test sets, and validate performance across lighting conditions, occupant types, and vehicle interiors

• Optimise and deploy models for real-time inference on embedded automotive hardware (e.g. quantisation, pruning, TensorRT)

• Combine model outputs with rule-based logic to detect safety events and rule violations with low false-alarm rates

• Apply privacy-by-design and secure-by-design principles to data handling and on-device processing, in line with GDPR and cybersecurity best practices

• Evaluate open-source and commercial models, write well-tested code, and work closely with Engineering, Product, and DevOps on technical direction

What do you need to succeed?

• 5+ years of software or ML engineering experience, including 2+ years developing and deploying computer vision models to production

• Strong Python skills and experience with PyTorch and computer vision libraries (e.g. OpenCV, torchvision, MMDetection/MMPose)

• Hands-on experience with object detection, pose estimation, and/or video action recognition (e.g. YOLO- or DETR-family detectors, transformer-based video models)

• Experience optimising models for edge deployment (e.g. ONNX, TensorRT, quantisation) on platforms such as NVIDIA Jetson or DRIVE Orin

• Solid ML engineering practices, including testing, CI/CD, experiment tracking, and model versioning (e.g. MLflow, Weights & Biases)

• Comfort making technical decisions and building systems without existing infrastructure, and explaining them clearly to a cross-functional team

• Degree in Computer Science, Electrical Engineering, Robotics, or a related field, or equivalent practical experience

Nice to have:

• Experience as an early or first ML hire

• Experience with in-cabin or driver monitoring, automotive, or robotics perception systems

• Experience with infrared imaging, large-scale labelling workflows, or synthetic data generation

• Familiarity with Euro NCAP occupant monitoring requirements, ISO 26262 / SOTIF, or GDPR in a data-heavy context

• Experience with Azure ML or cloud GPU training infrastructure

What’s in it for you?

Joining Verne means becoming part of a team shaping the future of autonomous mobility. Here’s what we offer:

• The chance to collaborate with experts from 20+ industries.

• Ownership of your projects, with the freedom to innovate and take initiative.

• A culture of transparency, feedback, and open communication.

• Comprehensive financial and wellbeing support designed to meet diverse needs.

• The excitement of working in a fast-growing startup where your ideas truly matter.

How can you apply?

If this role and our mission resonate with you, hit the Apply button! We typically review applications within 3-4 business days, but since we want to give each one the attention it deserves, please allow a little extra time if needed. Once we've had a chance to fully evaluate your submission, we’ll reach out with details about the next steps.

If shortlisted, here’s what to expect:

1. An initial HR interview focused on culture fit.

2. A technical interview with the Hiring Manager.

3. A final interview with C-level.

Not sure if this role is the right fit?

We’re always looking for creators who challenge the status quo and aren’t afraid to take the lead. If that’s you, explore our story - we might have other opportunities that spark your interest.

Verne (Project 3 Mobility) is a company comprised of people with different qualities and backgrounds, because we believe our differences make us stronger. That is why we evaluate qualified applicants fairly and equally, without regard to race, colour, religion, sex, sexual orientation, gender identity, national origin, disability, veteran status, age, familial status, and other legally protected characteristics. All applications will be considered in accordance with the regulations of personal data protection.